Papers

New on arXiv

Wed 6 Aug 2025

420 papers

Emergent time-keeping mechanisms in a deep reinforcement learning agent performing an interval timing taskDeep reinforcement learning methods → Deep RL algorithms and representations
Chain-of-Agents: End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RLLLM reasoning → Reasoning for agents, planning and code +1
StackPilot: Autonomous Function Agents for Scalable and Environment-Free Code ExecutionLLM agents and tool use → Software engineering coding agents
Sparse Attention across Multiple-context KV CacheEfficient LLM inference and serving → KV cache optimization
Training-Free Multimodal Large Language Model OrchestrationMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Guided Navigation in Knowledge-Dense Environments: Structured Semantic Exploration with Guidance GraphsLLM reasoning → Knowledge- and tool-augmented reasoning
An Audit and Analysis of LLM-Assisted Health Misinformation Jailbreaks Against LLMsLLM safety and security → Safety alignment, refusal and guardrails
Generating Feasible and Diverse Synthetic Populations Using Diffusion ModelsDiffusion models → Diffusion foundations and sampling
On the effectiveness of multimodal privileged knowledge distillation in two vision transformer based diagnostic applicationsKnowledge distillation → Distillation for medical imaging
A tutorial note on collecting simulated data for vision-language-action modelsVision-language-action models → Generalist VLA models and benchmarks
MambaITD: An Efficient Cross-Modal Mamba Network for Insider Threat DetectionState-space sequence models (S4, Mamba) → Time series forecasting with SSMs
Risk Analysis Techniques for Governed LLM-based Multi-Agent SystemsLLM agents and tool use → Agent threats, guardrails and governance
I Think, Therefore I Am Under-Qualified? A Benchmark for Evaluating Linguistic Shibboleth Detection in LLM Hiring EvaluationsLLM reasoning → Reasoning benchmarks and evaluation
RCR-Router: Efficient Role-Aware Context Routing for Multi-Agent LLM Systems with Structured MemoryLLM agents and tool use → General LLM agent frameworks and benchmarks
Retrieval-Augmented Foundation Models for Water Level Prediction in the EvergladesRetrieval-augmented generation → Time series and agricultural RAG
The Cosine Schedule is Fisher-Rao-Optimal for Masked Discrete Diffusion ModelsDiffusion models → Diffusion foundations and sampling
Sequence Aware SAC Control for Engine Fuel Consumption Optimization in Electrified PowertrainDeep reinforcement learning methods → RL for energy, networks and operations
VER-Bench: Evaluating MLLMs on Reasoning with Fine-Grained Visual EvidenceLLM reasoning → Multimodal reasoning
Large Language Models Reasoning Abilities Under Non-Ideal Conditions After RL-Fine-TuningLLM reasoning → Reinforcement learning for reasoning
Unified Flow Matching for Long Horizon Event ForecastingDiffusion models → Flow matching, time series and science
Charts-of-Thought: Enhancing LLM Visualization Literacy Through Structured Data ExtractionLLM reasoning → Multimodal reasoning
Persistent Instability in LLM's Personality Measurements: Effects of Scale, Reasoning, and Conversation HistoryLLM reasoning → Reasoning benchmarks and evaluation
Voost: A Unified and Scalable Diffusion Transformer for Bidirectional Virtual Try-On and Try-OffDiffusion models → 3D, point cloud and scene generation
Single-Step Reconstruction-Free Anomaly Detection and Segmentation via Diffusion ModelsDiffusion models → Flow matching, time series and science
RetinexDual: Retinex-based Dual Nature Approach for Generalized Ultra-High-Definition Image RestorationState-space sequence models (S4, Mamba) → Mamba for image restoration
Embedding Is (Almost) All You Need: Retrieval-Augmented Inference for Generalizable Genomic Prediction TasksRetrieval-augmented generation → Medical and biomedical RAG
AttriLens-Mol: Attribute Guided Reinforcement Learning for Molecular Property Prediction with Large Language ModelsLLM reasoning → Clinical and biomedical applications
A Foundational Multi-Modal Model for Few-Shot LearningMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Edge-Assisted Collaborative Fine-Tuning for Multi-User Personalized Artificial Intelligence Generated Content (AIGC)Diffusion models → Robustness, privacy and adversarial aspects
SEAgent: Self-Evolving Computer Use Agent with Autonomous Learning from ExperienceMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +2
Hop, Skip, and Overthink: Diagnosing Why Reasoning Models Fumble during Multi-Hop AnalysisLLM reasoning → Knowledge- and tool-augmented reasoning
FaST: Feature-aware Sampling and Tuning for Personalized Preference Alignment with Limited DataLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Achieving Precise and Reliable Locomotion with Differentiable Simulation-Based System IdentificationDeep reinforcement learning methods → RL for robot locomotion and manipulation
Robust Prompt Tuning for Vision-Language Models with Mild Semantic NoiseMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
HierarchicalPrune: Position-Aware Compression for Large-Scale Diffusion ModelsDiffusion models → Diffusion foundations and sampling
Composing Policy Gradients and Prompt Optimization for Language Model ProgramsLLM reasoning → Reinforcement learning for reasoning
LLM Collaboration With Multi-Agent Reinforcement LearningLLM agents and tool use → Reinforcement learning for LLM agents
EncQA: Benchmarking Vision-Language Models on Visual Encodings for ChartsMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
A Scalable Pretraining Framework for Link Prediction with Efficient AdaptationMixture-of-experts → MoE for time series and graphs
RoboTron-Sim: Improving Real-World Driving via Simulated Hard-CaseEnd-to-end autonomous driving → Generative simulation for driving
VirtLab: An AI-Powered System for Flexible, Customizable, and Large-scale Team SimulationsLLM agents and tool use → Social simulation and multi-agent debate
P-Aligner: Enabling Pre-Alignment of Language Models via Principled Instruction SynthesisLLM alignment / RLHF / preference optimisation → LLM preference optimization methods +1
FinMMR: Make Financial Numerical Reasoning More Multimodal, Comprehensive, and ChallengingLLM reasoning → Reasoning for agents, planning and code +1
Multitask Learning with Stochastic InterpolantsDiffusion models → Diffusion foundations and sampling
TURA: Tool-Augmented Unified Retrieval Agent for AI SearchRetrieval-augmented generation → Agentic RAG, agents and memory +1
ConfProBench: A Confidence Evaluation Benchmark for MLLM-Based Process JudgesLLM reasoning → Multimodal reasoning +1
Knowledge to Sight: Reasoning over Visual Attributes via Knowledge Decomposition for Abnormality GroundingLLM reasoning → Clinical and biomedical applications
Analyzing and Mitigating Object Hallucination: A Training Bias PerspectiveMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs +1
TopKD: Top-scaled Knowledge DistillationKnowledge distillation → General knowledge distillation methods
Balancing Stylization and Truth via Disentangled Representation SteeringMechanistic interpretability → Activation steering methods
RAIDX: A Retrieval-Augmented Generation and GRPO Reinforcement Learning Framework for Explainable Deepfake DetectionLLM reasoning → Multimodal reasoning +1
Causal Reflection with Language ModelsLLM reasoning → Causal reasoning in LLMs
QuantVSR: Low-Bit Post-Training Quantization for Real-World Video Super-ResolutionDiffusion models → Inverse problems and image restoration
OS Agents: A Survey on MLLM-based Agents for General Computing Devices UseLLM agents and tool use → GUI and mobile agents
TRAIL: Joint Inference and Refinement of Knowledge Graphs with Large Language ModelsLLM reasoning → Knowledge- and tool-augmented reasoning
FrEVL: Leveraging Frozen Pretrained Embeddings for Efficient Vision-Language UnderstandingMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
4DVD: Cascaded Dense-view Video Diffusion Model for High-quality 4D Content GenerationDiffusion models → 3D, point cloud and scene generation
CARD: A Cache-Assisted Parallel Speculative Decoding Framework via Query-and-Correct Paradigm for Accelerating LLM InferenceEfficient LLM inference and serving → Speculative decoding
From "Aha Moments" to Controllable Thinking: Toward Meta-Cognitive Reasoning in Large Reasoning Models via Decoupled Reasoning and ControlLLM reasoning → Test-time compute and efficient thinking
Boosting Visual Knowledge-Intensive Training for LVLMs Through Causality-Driven Visual Object CompletionMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Automatic LLM Red TeamingLLM safety and security → Prompt injection and agent security +1
StepFun-Formalizer: Unlocking the Autoformalization Potential of LLMs through Knowledge-Reasoning FusionLLM reasoning → Mathematical reasoning
GradSTL: Comprehensive Signal Temporal Logic for Neurosymbolic Reasoning and LearningLLM reasoning → Reasoning benchmarks and evaluation
Think Before You Segment: An Object-aware Reasoning Agent for Referring Audio-Visual SegmentationLLM reasoning → Multimodal reasoning
Thinking With Videos: Multimodal Tool-Augmented Reinforcement Learning for Long Video ReasoningLLM reasoning → Multimodal reasoning +1
Beyond Pixels: Exploring DOM Downsampling for LLM-Based Web AgentsLLM agents and tool use → Web, search, and research agents
FlexQ: Efficient Post-training INT6 Quantization for LLM Serving via Algorithm-System Co-DesignEfficient LLM inference and serving → KV cache quantization
GuirlVG: Incentivize GUI Visual Grounding via Empirical Exploration on Reinforcement LearningMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +2
TSPO: Temporal Sampling Policy Optimization for Long-form Video Language UnderstandingMultimodal LLMs / vision-language models → Video understanding with MLLMs
Multi-Marginal Stochastic Flow Matching for High-Dimensional Snapshot Data at Irregular Time PointsDiffusion models → Flow matching, time series and science
Chain of Questions: Guiding Multimodal Curiosity in Language ModelsLLM reasoning → Multimodal reasoning
GTPO and GRPO-S: Token and Sequence-Level Reward Shaping with Policy EntropyLLM reasoning → Reinforcement learning for reasoning
Deliberative Reasoning Network: An Uncertainty-Driven Paradigm for Belief-Tracked Inference with Pretrained Language ModelsLLM reasoning → Test-time compute and efficient thinking
Beyond the Leaderboard: Rethinking Medical Benchmarks for Large Language ModelsLLM reasoning → Clinical and biomedical applications
TempFlow-GRPO: When Timing Matters for GRPO in Flow ModelsLLM reasoning → Reinforcement learning for reasoning +2
Method-Based Reasoning for Large Language Models: Extraction, Reuse, and Continuous ImprovementLLM reasoning → Reasoning benchmarks and evaluation
Synthetic POMDPs to Challenge Memory-Augmented RL: Memory Demand Structure ModelingDeep reinforcement learning methods → Deep RL algorithms and representations
Prompt Injection Vulnerability of Consensus Generating Applications in Digital DemocracyLLM safety and security → Prompt injection and agent security
Enhancing Vision-Language Model Training with Reinforcement Learning in Synthetic Worlds for Real-World SuccessMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
A Few Words Can Distort Graphs: Knowledge Poisoning Attacks on Graph-based Retrieval-Augmented Generation of Large Language ModelsRetrieval-augmented generation → Privacy and security of RAG
ShoppingBench: A Real-World Intent-Grounded Shopping Benchmark for LLM-based AgentsLLM agents and tool use → Social simulation and multi-agent debate
KVSink: Understanding and Enhancing the Preservation of Attention Sinks in KV Cache Quantization for LLMsEfficient LLM inference and serving → KV cache quantization
Empowering Time Series Forecasting with LLM-AgentsLLM agents and tool use → Financial trading and forecasting agents
Intention Enhanced Diffusion Model for Multimodal Pedestrian Trajectory PredictionDiffusion models → Diffusion and flow policies for robotics
Continual Learning for VLMs: A Survey and Taxonomy Beyond ForgettingMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Symmetric Behavior Regularized Policy OptimizationOffline RL and imitation learning from data → Offline RL algorithms +1
Causal Reward Adjustment: Mitigating Reward Hacking in External Reasoning via Backdoor CorrectionLLM reasoning → Reinforcement learning for reasoning +2
DP-DocLDM: Differentially Private Document Image Generation using Latent Diffusion ModelsDiffusion models → Robustness, privacy and adversarial aspects
ReasoningGuard: Safeguarding Large Reasoning Models with Inference-time Safety Aha MomentsLLM reasoning → Reasoning for agents, planning and code +1
ViFP: A Framework for Visual False Positive Detection to Enhance Reasoning Reliability in VLMsLLM reasoning → Multimodal reasoning +1
Reasoning Beyond Labels: Measuring LLM Sentiment in Low-Resource, Culturally Nuanced ContextsLLM reasoning → Audio-language model reasoning
Gather and Trace: Rethinking Video TextVQA from an Instance-oriented PerspectiveLLM reasoning → Multimodal reasoning
Eliciting and Analyzing Emergent Misalignment in State-of-the-Art Large Language ModelsLLM safety and security → Safety alignment, refusal and guardrails
From Learning to Unlearning: Biomedical Security Protection in Multimodal Large Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
COPO: Causal-Oriented Policy Optimization for Hallucinations of MLLMsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
AD-FM: Multimodal LLMs for Anomaly Detection via Multi-Stage Reasoning and Fine-Grained Reward OptimizationLLM reasoning → Chain-of-thought reasoning for driving +1
STEMTOX: From Collaborative Tags to Fine-Grained Toxic Meme Detection via Entropy-Guided Multi-Task LearningLLM safety and security → Safety alignment, refusal and guardrails
Difficulty-Based Preference Data Selection by DPO Implicit Reward GapLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
IDCNet: Guided Video Diffusion for Metric-Consistent RGBD Scene Generation with Precise Camera ControlDiffusion models → 3D, point cloud and scene generation
COPO: Consistency-Aware Policy OptimizationLLM reasoning → Reinforcement learning for reasoning

All 420